Green computing or green IT, refers to environmentally sustainable computing or IT. In the article Harnessing Green IT: Principles and Practices, San Murugesan defines the field of green computing as "the study and practice of designing, manufacturing, using, and disposing of computers, servers, and associated subsystems—such as monitors, printers, storage devices, and networking and communications systems—efficiently and effectively with minimal or no impact on the environment."[1] The goals of green computing are similar to green chemistry; reduce the use of hazardous materials, maximize energy efficiency during the product's lifetime, and promote the recyclability or biodegradability of defunct products and factory waste. Research continues into key areas such as making the use of computers as energy-efficient as possible, and designing algorithms and systems for efficiency-related computer technologies.
Green Computing
mercredi 18 mai 2011
Performance per watt
In computing, performance per watt is a measure of the energy efficiency of a particular computer architecture or computer hardware. Literally, it measures the rate of computation that can be delivered by a computer for every watt of power consumed.
The performance and power consumption metrics used depend on the definition; reasonable measures of performance are FLOPS, MIPS, or the score for any performance benchmark. Several measures of power usage may be employed, depending on the purposes of the metric; for example, a metric might only consider the electrical power delivered to a machine directly, while another might include all power necessary to run a computer, such as cooling and monitoring systems. Often the power is the average power used while running the benchmark, but sometimes other measures of power usage may be employed (e.g. peak power, idle power.)
For example, the early UNIVAC I computer performed approximately 0.015 operations per watt-second (performing 1,905 operations per second, while consuming 125 kW).
Most of the power a computer uses is converted into heat, so a system that takes fewer watts to do a job will require less cooling to maintain a given operating temperature. Reduced cooling demands makes it easier to quiet a computer. Lower energy consumption can also make it less costly to run, and reduce the environmental impact of powering the computer (see green computing). If installed where there is limited climate control, a lower power computer will operate at a lower temperature, which may make it more reliable. In a climate controlled environment, reductions in direct power use may also create savings in climate control energy.
Computing energy consumption is sometimes also measured by reporting the energy required to run a particular benchmark, for instance EEMBC EnergyBench. Energy consumption figures for a standard workload may make it easier to judge the effect of an improvement in energy efficiency, just as the use of L/100km is easier than reciprocal measures (such as miles per gallon) when judging impact of automotive fuel economy.
Performance (in operations/second) per watt can also be written as operations/watt-second, or operations/joule, since 1 watt = 1 joule/second.
The performance and power consumption metrics used depend on the definition; reasonable measures of performance are FLOPS, MIPS, or the score for any performance benchmark. Several measures of power usage may be employed, depending on the purposes of the metric; for example, a metric might only consider the electrical power delivered to a machine directly, while another might include all power necessary to run a computer, such as cooling and monitoring systems. Often the power is the average power used while running the benchmark, but sometimes other measures of power usage may be employed (e.g. peak power, idle power.)
For example, the early UNIVAC I computer performed approximately 0.015 operations per watt-second (performing 1,905 operations per second, while consuming 125 kW).
Most of the power a computer uses is converted into heat, so a system that takes fewer watts to do a job will require less cooling to maintain a given operating temperature. Reduced cooling demands makes it easier to quiet a computer. Lower energy consumption can also make it less costly to run, and reduce the environmental impact of powering the computer (see green computing). If installed where there is limited climate control, a lower power computer will operate at a lower temperature, which may make it more reliable. In a climate controlled environment, reductions in direct power use may also create savings in climate control energy.
Computing energy consumption is sometimes also measured by reporting the energy required to run a particular benchmark, for instance EEMBC EnergyBench. Energy consumption figures for a standard workload may make it easier to judge the effect of an improvement in energy efficiency, just as the use of L/100km is easier than reciprocal measures (such as miles per gallon) when judging impact of automotive fuel economy.
Performance (in operations/second) per watt can also be written as operations/watt-second, or operations/joule, since 1 watt = 1 joule/second.
Servers and data centers power management
Servers and data centers account for 23% of IT energy demand.[3] Data centers are a point source energy demand where a lot of focus has been on input control to reduce the power demand of cooling systems, and to make power inputs to servers themselves more efficient.[clarification needed]
As computing hardware becomes smaller and less expensive, energy costs constitute a larger portion of server or data center costs.[9]
Server and data center systems tend to be designed with significant computational redundancy. Typically, an individual server will only operate at around 18% of its capacity.[citation needed] The reasons for this are largely historical, and with current technology, this level of redundancy is not required.[citation needed]
This feature of data centers and servers allows major energy efficiency gains to be made through optimisation of servers. This is typically done by doing diagnostic tests on individual servers and developing a model for a data center’s energy demand using these measurements.
By analysing every server in a data centre, server power management software can identify servers that can be removed. It also enables servers to be virtualized, processes to be consolidated to a smaller number of servers, and servers with a predictable cyclical power demand to be fully powered down when not in use. Active power management features are also included which put remaining servers into their lowest power state that allows instant wake-up on demand when required.
This approach to server power management has the potential to make very large energy savings in data centers and server rooms.
Energy efficiency benchmarks, such as SPECpower, or specifications, like Average CPU power, can be used to comparing server efficiency and performance per watt.
Siting of data centers can increase or reduce their energy use. Siting in areas where climate allows air-cooling and lots of renewable electricity is available nearby can reduce the energy used and resulting environmental effects.
As computing hardware becomes smaller and less expensive, energy costs constitute a larger portion of server or data center costs.[9]
Server and data center systems tend to be designed with significant computational redundancy. Typically, an individual server will only operate at around 18% of its capacity.[citation needed] The reasons for this are largely historical, and with current technology, this level of redundancy is not required.[citation needed]
This feature of data centers and servers allows major energy efficiency gains to be made through optimisation of servers. This is typically done by doing diagnostic tests on individual servers and developing a model for a data center’s energy demand using these measurements.
By analysing every server in a data centre, server power management software can identify servers that can be removed. It also enables servers to be virtualized, processes to be consolidated to a smaller number of servers, and servers with a predictable cyclical power demand to be fully powered down when not in use. Active power management features are also included which put remaining servers into their lowest power state that allows instant wake-up on demand when required.
This approach to server power management has the potential to make very large energy savings in data centers and server rooms.
Energy efficiency benchmarks, such as SPECpower, or specifications, like Average CPU power, can be used to comparing server efficiency and performance per watt.
Siting of data centers can increase or reduce their energy use. Siting in areas where climate allows air-cooling and lots of renewable electricity is available nearby can reduce the energy used and resulting environmental effects.
Energy Efficient Ethernet
Energy Efficient Ethernet (IEEE 802.3az) could reduce the energy use of networking equipment. In 2005, all the network-interface controllers in the United States (in computers, switches, and routers) used an estimated 5.3 terawatt-hours of electricity.[7] According to a researcher at the Lawrence Berkeley Laboratory, Energy Efficient Ethernet could save an estimated $450 million a year in energy costs in the U.S.[8] With most of the savings from home computers ($200 million), and offices ($170 million), and the remaining $80 million from data centers.[8] Energy efficient Ethernet saves energy by allowing network links to either go into a low power sleep mode or run at a slower rate when there is no data. It also defines lower power signaling for use on higher quality cables.
PC power management
Research shows that in the US, 50% of PCs are left on overnight, resulting in an estimated annual energy waste of 28.8 billion KWH, and a cost of $2.8 billion to the economy. User behaviour is slightly different in Europe, with approximately 28% of PCs being left on overnight in the UK, resulting in an estimated energy loss of 2.5 billion KWH, costing £300 million. In Germany, with approximately 30% of PCs left on overnight, it is estimated 4.8 billion KWH of energy are wasted each year, costing €919 million[4]
User behaviour is a big factor in energy waste by PCs, through people not turning off their PCs when they leave work, but network processes also compound the problem.
Managing energy use by PCs attached to IT networks is difficult due to network processes making a PC’s internal power profiles ineffective. Because processes on the network are constantly providing inputs to the PC, the machine fails to recognise that there is no user input, and fails to go into its low-power mode.
PC power management is a classic problem of distributed demand management, where individual devices which have a relatively low power demand result in a very large cumulative energy demand.
Addressing this problem requires specialised power management software. These tools are highly effective at cutting power demand by PCs,[5] saving of 40% on average can be made using this software. Typical savings per PC are $35 and 200 kg CO2 per PC per year.
User behaviour is a big factor in energy waste by PCs, through people not turning off their PCs when they leave work, but network processes also compound the problem.
Managing energy use by PCs attached to IT networks is difficult due to network processes making a PC’s internal power profiles ineffective. Because processes on the network are constantly providing inputs to the PC, the machine fails to recognise that there is no user input, and fails to go into its low-power mode.
PC power management is a classic problem of distributed demand management, where individual devices which have a relatively low power demand result in a very large cumulative energy demand.
Addressing this problem requires specialised power management software. These tools are highly effective at cutting power demand by PCs,[5] saving of 40% on average can be made using this software. Typical savings per PC are $35 and 200 kg CO2 per PC per year.
Green computing and the industry
![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhdsCLgnmVkFfb9vbZh7aExv4JpvCkr988JAh48TE64kC3B-9sESLTf4Dq1MWpuuk-ICcdIkDkTuF9YI_wyIq_mKZluNpYdP7jH7jlivK3SvK6eMiyYaqMx8B-MJwTmAUX5KLSWiwinsiE/s1600/k4876617.jpg)
The Green Electronics Council offers the Electronic Product Environmental Assessment Tool (EPEAT) to assist in the purchase of "greener" computing systems. The Council evaluates computing equipment on 51 criteria - 23 required and 28 optional - that measure a product's efficiency and sustainability attributes. Products are rated Gold, Silver or Bronze depending on how many optional criteria they meet. On 2007-01-24, President George W. Bush issued Executive Order 13423, which requires all United States Federal agencies to use EPEAT when purchasing computer systems.[12][13]
The Green Grid is a global consortium dedicated to advancing energy efficiency in data centers and business computing ecosystems. It was founded in February 2007 by several key companies in the industry – AMD, APC, Dell, HP, IBM, Intel, Microsoft, Rackable Systems, SprayCool, Sun Microsystems and VMware. The Green Grid has since grown to hundreds of members, including end users and government organizations, all focused on improving data center efficiency.
The Green500 list rates supercomputers by energy efficiency (megaflops/watt, encouraging a focus on efficiency rather than absolute performance.
Green Comm Challenge is an organization that promotes the development of energy conservation technology and practices in the field of Information and Communications Technology (ICT).
The Transaction Processing Performance Council(TPC) Energy specification augments the existing TPC benchmarks by allowing for optional publications of energy metrics alongside their performance results.[14]
The SPEC Power is the first industry standard benchmark that measures power consumption in relation to performance for server-class computers.
Trashware
Trashware in North America or Totally reconditioned hardware in the UK and Ireland is computer equipment that is assembled from old hardware, using cleaned and checked parts from different computers, for use by disadvantaged people to bridge the digital divide.
Trashware, with its social aims, is different from retrocomputing, which has only cultural and recreational purposes.
Trashware, with its social aims, is different from retrocomputing, which has only cultural and recreational purposes.
Computers being collected for recycling at a pickup event in Olympia, Washington, United States. |
Inscription à :
Articles (Atom)